ed.vul.lab - people
...    e. vul lab on cognition & inference

lab members

contact us at evullab@gmail.com

ed vul - was raised very close to Chernobyl, which may explain a few things. During undergrad he worked with Hal Pashler and Don MacLeod here at UCSD, he got a PhD with Nancy Kanwisher and Josh Tenenbaum at MIT, and then came back to UCSD for a sweet faculty gig.
He now gets to work with this awesome lot of people and muse about puzzles such as: How do people carry out approximately rational inference despite their limited cognitive resources? How should they allocate those limited resources? How do we formalize the interaction of our rich world knowledge with our cognitive capacity limitations to predict real-world behavior?
Personal website

nisheeth srivastava - As good Bayesian cognitive scientists, we often operate from the point of view that the way animals' thoughts about the world evolve can be accurately described as statistically optimal induction. However, being statistically optimal can be very hard biophysical work for the animal. My primary interest is figuring out how animals trade off information for energy and vice versa by developing and testing cognitive theories that are sensitive to this trade-off.
Personal website

kevin smith - I am interested in understanding how people retrieve and use background knowledge about the world, and how this combines with the inherent uncertainty about the world to influence our predictions and decisions. I use various computational modeling techniques to match theories about how people are retrieving and processing information, and how people behave in experimental tasks.
I currently have two research focuses: how people understand how objects move about and interact in the world (intuitive physics), and how people combine multiple constraints to search through semantic memory.
Personal website

tim lew - I'm a fourth year grad student (as of Fall 2015). I did undergrad at the University of Pennsylvania (C'2012) where I worked under Mike Kahana investigating behavioral models of human spatial memory.

My current research focuses primarily on how memory uses structure in the world to encode and recall information. For example, remembering the locations of people in a crowd becomes much easier if I infer the crowd is arranged in a line or cluster. Visual memory appears to rely on such statistical structure frequently as a means of efficiently and accurately representing the features of objects.

I also apparently like to use my free time to convert my experiment code into video games about my labmates (for science purposes, of course).

Personal website

drew walker -I am a fourth year graduate student. I did my undergraduate work here at UCSD working with John Wixted, Laura Mickes and Timothy Rickard . My current research focuses on examining a cognitive explanation of "the cheerleader effect", the notion that people are more attractive when in a group. Specifically, I am investigating the idea that it is an interplay of ensemble coding in the visual system, and the attractivness of averaged faces that produces this effect.

lambda(λ) - Lambda's primary research interest is predicting and manipulating whether, when, and where people drop food on the floor.

banjo - Banjo makes important contributions to many fields. And fire hydrants. And lamp posts.


randy tran - I am a second year grad student being advised by Hal Pashler . I am interested in learning, education and inference-forming.

lab alumni

cory rieth - I collect behavioral data and develop computational models of human perception and attention to understand the dynamic interaction between the mind and the world. For example: How do expectations about the environment influence perception? How do we segregate and select information in time? How do we learn about the structure of the world?
My modeling approach primarily treats perception and attention as the result of approximate rational inference, meaning that I think that people usually do what they should be doing given their typical environment (at least more often than we give them credit for) and with approximations to rational inference, e.g. sampling.
Personal website